Arena: the environment repository, and redaction-pressure

An environment is an eval you take the gradient of, so it carries obligations a
benchmark does not: a held-out slice, a counterweighted reward, and a demonstrated
0.0 floor and 1.0 ceiling. README states the three as house rules.

redaction-pressure is Forge's redact stage as a task. The first reward design scored
survivors — secrets caught, decoys kept, prose kept — and an empty ruleset tied the
best real attempt at 0.500, because destroying nothing keeps everything. Counting
removals instead puts inaction at 0.000, shredding at 0.380, and an oracle at 1.000.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-08-19 00:09:31 -07:00
co-authored by Claude Opus 5
commit c28d864766
10 changed files with 4208 additions and 0 deletions
@@ -0,0 +1,13 @@
[project]
name = "redaction-pressure"
version = "0.1.0"
description = "redaction-pressure — write redaction rules graded on records the author never saw."
requires-python = ">=3.11"
dependencies = ["verifiers", "regex"]
[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"
[tool.hatch.build.targets.wheel]
packages = ["redaction_pressure"]